Canada Uses Claude Four Times More Than Expected — and Workforce Mix, Not Income, Explains It
TRANSMISSION RECEIVED · PLANET MAPLE-1 · COORDS [-0.34, 0.62]
Anthropic’s latest Economic Index installment turns the lens on a single country — Canada — and finds something counterintuitive: per capita, Canadians use Claude at more than four times the rate their population would predict, second only to the United States. But the real finding isn’t the headline. Within Canada, adoption tracks the shape of the local workforce, not the size of the local paycheck.
A geographic x-ray of adoption
The Index leans on a metric called the Anthropic AI Usage Index (AUI): usage share adjusted for working-age population, where 1.0 means a place uses Claude exactly as much as its population would predict. Canada scores 4.4 — over 4x expected. It accounts for 2.6% of global Claude.ai traffic, ranking eighth by volume but second by AUI, behind only the US.
Cross-country, income explains a lot: advanced economies use Claude more. But Canada overshoots even its income prediction, which the authors attribute to a highly educated workforce and proximity to the US technology frontier. So far, a rich-country-uses-more-AI story.
The within-Canada data breaks that story. Usage is geographically concentrated: Ontario alone accounts for 43.9% of conversations, and four provinces (Ontario, Quebec, British Columbia, Alberta) make up roughly 94%. Adjusted for population, British Columbia sits at 1.4x expected, Ontario at 1.1x, while Newfoundland and Labrador sits at 0.2x. The twist: provincial income per capita does not explain this gap. What does is industrial composition — regions with larger professional, scientific, and technical services sectors use Claude systematically more. Adoption diffuses along the grain of a region’s knowledge-work density, not its wealth.
Usage patterns, by contrast, are remarkably uniform. Personal use (health lookups, product research, recipes) runs 44–51% in every province; work (troubleshooting, drafting emails, building apps) runs 34–40%; coursework 13–18%. What varies is the distinctive use case. Translation requests track public-administration employment shares — New Brunswick, Nova Scotia, and Quebec, the provinces with the most bilingual-federal workforces, also devote the largest share of conversations to translation. Document translation is the single most distinctive Canadian use case relative to Anglosphere peers. More broadly, Canadian usage skews academic and early-career: coursework, coding help, and resume drafting are overrepresented; professional communication and routine personal tasks are underrepresented.
The throughline: in high-income countries, adoption is shaped less by income than by how well model capabilities match the structure of the local economy.
Why marketers should care
AI-literate audiences are not evenly spread — they concentrate where knowledge-work sectors concentrate, and they use tools in ways that mirror local conditions. Canada’s translation skew is the clearest example: a national policy (official bilingualism) shows up as a behavioral signal in tool usage. That matters for market sizing, localization priorities, and which use cases to lead with in a given region. The early-career tilt also signals that the next cohort of professionals and decision-makers is arriving AI-native — not a future audience, a present one.
How to use it
- Size by workforce, not GDP. When prioritizing markets or regions, weight by knowledge-service sector density — the Index suggests that’s the better predictor of AI-tool adoption.
- Localize for the distinctive use case. Where translation over-indexes (as in Quebec), lead with language-localized workflows rather than English-first templates.
- Meet early-career users where they are. Coursework, coding, and resume help are entry points — relevant for campus recruiting, employer brand, and product surfaces aimed at junior talent.
// END OF LOG